Re: [SAGEdev] Error building sage in Solaris 10 with gcc 4.0.2
David Joyner
Thanks for your report but it was sent to the wrong list.
(The sf list is "closed". I'm forwarding this to the correct list.)
- David Joyner
On Nov 14, 2007 8:19 AM, Klas Heggemann <kl...@nada.kth.se> wrote:
>
> This fails in the building of one of the libraries:
>
> SunOS sgray.nada.kth.se 5.10 Generic_127111-02 sun4u sparc
> SUNW,Sun-Fire-280R
> ****************************************************
> ****************************************************
> GCC Version
> gcc -v
> Using built-in specs.
> Target: sparc-sun-solaris2.10
> Configured with: ../gcc-4.0.2/configure --prefix=/pkg/gcc/4.0.2/os
> Thread model: posix
> gcc version 4.0.2
> ****************************************************
> make[2]: Entering directory
> `/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/spkg/build/flint-0.2.p4/src'
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c mpn_extras.c -o
> mpn_extras.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c Z.c -o Z.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c memory-manager.c -o
> memory-manager.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c Z_mpn.c -o Z_mpn.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c ZmodF.c -o ZmodF.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c ZmodF_mul.c -o
> ZmodF_mul.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c ZmodF_mul-tuning.c
> -o ZmodF_mul-tuning.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c fmpz.c -o fmpz.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c fmpz_poly.c -o
> fmpz_poly.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c mpz_poly-tuning.c
> -o mpz_poly-tuning.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c mpz_poly.c -o
> mpz_poly.o
> gcc -std=c99 -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include/
> -I/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/local/include
> -fexpensive-optimizations -fPIC -funroll-loops -O3 -c ZmodF_poly.c -o
> ZmodF_poly.o
> ZmodF_poly.c: In function 'ZmodF_poly_bit_pack_mpn':
> ZmodF_poly.c:217: error: 'u_int16_t' undeclared (first use in this
> function)
> ZmodF_poly.c:217: error: (Each undeclared identifier is reported only once
> ZmodF_poly.c:217: error: for each function it appears in.)
> ZmodF_poly.c:217: error: syntax error before 'lower'
> ZmodF_poly.c:269: error: 'lower' undeclared (first use in this function)
> ZmodF_poly.c:269: error: syntax error before 'temp'
> make[2]: *** [ZmodF_poly.o] Error 1
> make[2]: Leaving directory
> `/afs/.nada.kth.se/pkg/sage/src/sage-2.8.12/spkg/build/flint-0.2.p4/src'
>
>
> This is due to the fact that there is no typedef for u_int_t in include
> files.
>
> Adding this definition in some file makes sage build OK.
>
>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems? Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >> http://get.splunk.com/
> _______________________________________________
> sage-devel mailing list
> sage-...@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/sage-devel
>
>
mabshoff
On Nov 14, 3:59 pm, "David Joyner" <wdjoy...@gmail.com> wrote:
> Klas:
> Thanks for your report but it was sent to the wrong list.
> (The sf list is "closed". I'm forwarding this to the correct list.)
:)
> - David Joyner
mmh, I remember fixing that. Maybe it snuck back in or Bill has fixed
it upstream only.
Either way I will check this out on my Sun 10 box. Last time I checked
I got everything to build except cvxopt (due to missing complex.h, but
that might be fixed now, too), but some of the doctests failed upon
startup.
Klas, if you have the time: Could you run ./sage -testall and report
back please? Apologies if I got the wrong Klas Heggemann :)
Cheers,
Michael
>
> > Adding this definition in some file makes sage build OK.
>
> > -------------------------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc.
> > Still grepping through log files to find problems? Stop.
> > Now Search log events and configuration files using AJAX and a browser.
> > Download your FREE copy of Splunk now >>http://get.splunk.com/
> > _______________________________________________
> > sage-devel mailing list
> > sage-de...@lists.sourceforge.net
> >https://lists.sourceforge.net/lists/listinfo/sage-devel
Michael Abshoff
> mabshoff wrote:
>>
Hi Klas,
>> mmh, I remember fixing that. Maybe it snuck back in or Bill has fixed
>> it upstream only.
I will open a ticket for this in a minute or two so it doesn't get lost.
>>
>> Either way I will check this out on my Sun 10 box. Last time I checked
>> I got everything to build except cvxopt (due to missing complex.h, but
>> that might be fixed now, too), but some of the doctests failed upon
>> startup.
>>
>> Klas, if you have the time: Could you run ./sage -testall and report
>> back please? Apologies if I got the wrong Klas Heggemann :)
>>
>
> You did...
>
> However the build does not seem to be functional:
>
> sgray:sage-2.8.12>./sage
> ----------------------------------------------------------------------
> | SAGE Version 2.8.12, Release Date: 2007-11-06 |
> | Type notebook() for the GUI, and license() for information. |
> ----------------------------------------------------------------------
> /usr/bin/env: No such file or directory
> /usr/bin/env: No such file or directory
Mmh, which shell do you use? Sage only works correctly with bash at the
moment, but I thought that Solaris 10 at least shipped with a decent
enough bash to run Sage. Otherwise I need to take a deeper look if it
isn't the bash :(
>
> It seems to try to locate some programs that is not avaliable,
> even using the same PATH as during the build.
>
> sgray:sage-2.8.12>./sage -testall
> This script will run the unit tests for DSage
> /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-dsage-trial:
> trial: not found
> However, an ERROR occurred in the Distributed SAGE unit tests.
> Testing SAGE documentation
> Testing SAGE tutorial
> /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-maketest:
> /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/devel/doc/tut: does not exist
>
>
>
Cheers,
Michael
Michael Abshoff
>
> On 15 nov 2007, at 16.45, Michael Abshoff wrote:
>>>
>>>
>>> However the build does not seem to be functional:
>>>
>>> sgray:sage-2.8.12>./sage
>>> ----------------------------------------------------------------------
>>> | SAGE Version 2.8.12, Release Date:
>>> 2007-11-06 |
>>> | Type notebook() for the GUI, and license() for
>>> information. |
>>> ----------------------------------------------------------------------
>>> /usr/bin/env: No such file or directory
>>> /usr/bin/env: No such file or directory
>>
>> Mmh, which shell do you use? Sage only works correctly with bash at
>> the
>> moment, but I thought that Solaris 10 at least shipped with a decent
>> enough bash to run Sage. Otherwise I need to take a deeper look if it
>> isn't the bash :(
>
> But please note that on Solaris /bin/sh is not bash, so
> if scripts start with
>
> #!/bin/sh
>
> some things may not work as expected.
>
> From what I cas see, most scripts in local/bin ar #!/bin/sh-scripts.
>
> If I change these to start with #!/bin/bash, things look different:
>
Yep, this is now ticket #1179
http://www.sagetrac.org/sage_trac/ticket/1179
On our Solaris test box we replaced /bin/sh by /bin/bash because without
that change at least clisp wouldn't build and there were some other
issues.
> sgray:sage-2.8.12>./sage -testall
> This script will run the unit tests for DSage
> /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-dsage-trial:
> line 17: trial: command not found
> However, an ERROR occurred in the Distributed SAGE unit tests.
> Testing SAGE documentation
> Testing SAGE tutorial
> /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-maketest:
> line 18: cd: /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/devel/doc/tut:
> No such file or directory
> ERROR: File ./tut.tex is missing
> exit code: 1
>
> ----------------------------------------------------------------------
> The following tests failed:
>
>
> ./tut.tex
> Total time for all tests: 0.0 seconds
> Testing SAGE programming guide
> /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-maketest:
> line 22: cd: /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/devel/doc/prog:
> No such file or directory
> ERROR: File ./prog.tex is missing
> exit code: 1
>
> ----------------------------------------------------------------------
> The following tests failed:
>
>
> ./prog.tex
> Total time for all tests: 0.0 seconds
> Testing SAGE constructions guide
> /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-maketest:
> line 26: cd: /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/devel/doc/
> const: No such file or directory
> ERROR: File ./const.tex is missing
> exit code: 1
>
> ----------------------------------------------------------------------
> The following tests failed:
>
>
> ./const.tex
> Total time for all tests: 0.0 seconds
> ERROR: File ./devel/sage/sage is missing
> exit code: 1
>
> ----------------------------------------------------------------------
> The following tests failed:
>
>
> ./devel/sage/sage
> Total time for all tests: 0.0 seconds
> However, an ERROR occurred in the Distributed SAGE unit tests.
> Please see /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/tmp/test.log for
> the complete log from this test.
It looks like the build failed somehow. I see that you use afs, so that
adds another variable to the mix. Could you gzip the install.log and put
it up somewhere so that I can have a look?
Cheers,
Michael
William Stein
<Michael...@fsmath.mathematik.uni-dortmund.de> wrote:
> > sgray:sage-2.8.12>./sage
> > ----------------------------------------------------------------------
> > | SAGE Version 2.8.12, Release Date: 2007-11-06 |
> > | Type notebook() for the GUI, and license() for information. |
> > ----------------------------------------------------------------------
> > /usr/bin/env: No such file or directory
> > /usr/bin/env: No such file or directory
>
> Mmh, which shell do you use? Sage only works correctly with bash at the
> moment, but I thought that Solaris 10 at least shipped with a decent
> enough bash to run Sage. Otherwise I need to take a deeper look if it
> isn't the bash :(
>
I don't know if his problem is a bash problem or not. All over the
place in sage we use
#!/usr/bin/env foobar
to run a script using the program foobar in the user's path.
If /usr/bin/env doesn't exist this won't work. Where is your env command?
What is the output of
which env
Could you put a symlink from your env to /usr/bin/env?
William
> >
> > It seems to try to locate some programs that is not avaliable,
> > even using the same PATH as during the build.
> >
> > sgray:sage-2.8.12>./sage -testall
> > This script will run the unit tests for DSage
> > /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-dsage-trial:
> > trial: not found
> > However, an ERROR occurred in the Distributed SAGE unit tests.
> > Testing SAGE documentation
> > Testing SAGE tutorial
> > /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-maketest:
> > /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/devel/doc/tut: does not exist
> >
> >
> >
>
> Cheers,
>
> Michael
>
>
>
> >
>
--
William Stein
Associate Professor of Mathematics
University of Washington
http://wstein.org
William Stein
<Michael...@fsmath.mathematik.uni-dortmund.de> wrote:
> >
> > sgray:sage-2.8.12>./sage -testall
> > This script will run the unit tests for DSage
> > /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-dsage-trial:
> > line 17: trial: command not found
> > However, an ERROR occurred in the Distributed SAGE unit tests.
> > Testing SAGE documentation
> > Testing SAGE tutorial
> > /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/local/bin/sage-maketest:
> > line 18: cd: /afs/nada.kth.se/pkg/sage/src/sage-2.8.12/devel/doc/tut:
> > No such file or directory
> > ERROR: File ./tut.tex is missing
> > exit code: 1
> >
That strongly suggests the documentation package wasn't installed.
I wonder to what extent your build of Sage (on solaris) actually
succeeded.
William